Regulations aimed at increasing low‐income Americans’ access to credit are getting a long‐overdue revamp. Two of the three agencies responsible for enforcement of the Community Reinvestment Act issued a proposal earlier this month to change the way they assess how banks lend to underserved communities. The proposal is modest, but it includes some important changes to CRA regulations that will focus on lending to low‐income households and recognize that banks are increasingly going branchless.
The CRA, enacted in 1977, applies to banks but not credit unions or fintech lenders. It requires banking regulators — the Office of the Comptroller of the Currency, the Federal Deposit Insurance Corporation and the Federal Reserve Board — to ensure banks “meet the credit needs” of the communities where they operate, without sacrificing bank safety and soundness. Regulators give banks a rating based on their performance. If banks do not perform well, regulators may block their future expansion and merger.
When the CRA came into being, competition between banks was limited: most states prohibited branching and no states allowed out‐of‐state banks to enter their markets. The Fed also capped interest on deposits, giving banks cheap access to funds. These barriers to competition have gradually disappeared since the 1980s, ushering in rapid consolidation, a near‐doubling of the number of bank offices, and a wider set of banking options for consumers.
At the same time, non‐banks such as Quicken and Kabbage have taken up a growing share of the mortgage and small‐business lending markets on which the CRA focuses. These non‐banks often lend to low‐income communities as much as or more than banks.
It has long been time to update CRA regulations to reflect these structural changes to the U.S. banking landscape. Yet the CRA has not undergone meaningful change for nearly 25 years. That’s why the OCC and FDIC reform proposal, without being ambitious, can better address the credit needs of vulnerable households.
For example, under the proposal, loans to high‐income borrowers would no longer earn banks CRA points. By counting both loans to low‐income borrowers and loans made in low‐income areas toward their CRA evaluations, regulators presently reward banks for extending mortgages to prosperous professionals who do not need help from the government. My research shows that, from 2012 to 2017, between 65 and 70 percent of CRA‐eligible mortgages in the District of Columbia went to high‐income residents of low‐income areas. Furthermore, there was a strong association between CRA‐eligible mortgage lending and declines in the minority share of residents of individual D.C. neighborhoods.
Gentrification is often inevitable, and it can have a very positive impact on neighborhoods and their historic residents, especially when zoning laws allow housing supply to respond effectively to rising demand. But gentrification neither needs nor deserves help from government policy.
The proposal also takes account of lending in areas where banks have substantial activity but no physical presence. Existing CRA rules evaluate banks only where they operate branches or ATMs. But the Internet has made it possible for banks to collect deposits and make loans without a physical presence. Digital banking far away from branches is now the growth strategy of some established banks as well as aspiring entrants. Regulation should recognize this change.
Under the OCC and FDIC’s proposal, banks that receive more than 50 percent of their deposits from outside the places where they have branches and ATMs will be assessed, and receive CRA points for, activity in the areas where more than 5 percent of their digital deposits originate. This change will not only better reflect the scope of a modern bank’s activity, but it will discourage banks from concentrating their CRA‐eligible lending in places that do not need it.
For example, Salt Lake City is a preferred headquarters location for nationally active internet banks. But it makes no sense to evaluate banks whose activity is national for their activity in Salt Lake only. Furthermore, Salt Lake may need CRA lending less than other parts of the country. The proposed change is therefore both more efficient and more equitable.
Not all of the proposed changes are desirable. For example, the various ratios proposed for quantifying banks’ CRA performance come dangerously close to prescriptive lending quotas. While putting numbers on banks’ lending activity can make compliance easier and increase the consistency of CRA evaluations, regulators simply lack the knowledge to know the right amount of lending for each bank in each part of the country. Indeed, one of the virtues of our market‐based banking system is that it collects funds from savers and lends them to qualified borrowers, regardless of where they live. Putting regulators in charge of credit allocation is not good for efficiency or financial stability.
Despite the proposal’s shortcomings, it is a shame that the Fed did not sign up to it. As the smallest CRA regulator, responsible for just 16 percent of banks by assets, the Fed is not expected to take a leading role in its review. But by failing to signal its support, the Fed risks fomenting inconsistency in CRA enforcement, which can only harm banks and the achievement of the CRA’s goals. As the proposal gives all institutions with assets below $500 million, comprising 72 percent of U.S. banks, the right to opt out of the proposed changes, the Fed’s reticence is still more baffling.
The OCC’s and FDIC’s CRA reform proposal will not revolutionize the enforcement of that 42‐year‐old law, but it will focus CRA activity on truly underserved borrowers and acknowledge the banking landscape of 2020. It very much deserves a hearing.